Experiment / E2D4OAB0KSingle-Cell MPRA (scMPRA / sc-lentiMPRA)

R1-scMPRA: single-cell enhancer activity

Iterative deep learning-design of human enhancers exploits condensed sequence grammar to achieve cell type-specificity

A single-cell MPRA measurement of the R1-MPRA enhancer library in a mixed HepG2/K562 population. Sparse enhancer and transcriptome matrices were pseudobulked after marker-based cell-state assignment to provide enhancer activity summaries for HepG2-like and K562-like cells.

Download processed table (CSV)Raw source data API documentation

Processed tables are specific to each experiment. Column names, units, measurements, and table structure are not standardized across the database. Check this experiment’s column definitions and quality-control notes before comparing or combining data.

Organism: Human
Taxonomy ID: NCBITaxon:9606
Biosample: CVCL:0027;CVCL:0004
Reference genome: GRCh38
Design focus: Synthetic / Motif-focused
Region of interest: Not reported / not applicable

Perturbation & assay details

Basal / untreated mixed-cell culture

The GEO submission provides sparse H5 matrices for enhancer reporter UMIs and transcriptome UMIs. Cells were assigned to HepG2-like or K562-like states using log-normalized canonical marker modules, then enhancer counts were aggregated and normalized by transcriptome UMI totals.

Processed data

50 rows per page. Click a cell to inspect its full value.

Visible columns (26 of 26)

Row
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50

Page 1 · 50 rows · More results available

Filters apply to this table only. The CSV download contains the complete processed table; filtered rows are available through the API.

Column dictionary · 26 definitions

element_id: Unique identifier for the tested enhancer element.
enhancer_sequence: DNA sequence assayed as the reporter enhancer.
sequence_length_bp: Length of the tested enhancer sequence in base pairs.
barcode_sequences: Semicolon-separated reporter barcode sequences associated with the element.
barcode_count: Number of barcodes associated with the element.
design_target_cell_type: Design target cell type.
design_model: Design model.
design_generator: Design generator.
design_model_type: Design model type.
dataset_split: Dataset split.
model_predicted_log2_H2K: Model-predicted HepG2-versus-K562 differential activity.
author_log2FoldChange_HepG2: Author log2foldchange hepg2.
author_log2FoldChange_K562: Author log2foldchange k562.
author_log2FoldChange_H2K: Authors’ DESeq2 HepG2-versus-K562 log2 fold-change or differential activity statistic.
hepg2_cell_count: Hepg2 cell count.
k562_cell_count: K562 cell count.
hepg2_mpra_umi_total: Hepg2 mpra umi total.
k562_mpra_umi_total: K562 mpra umi total.
hepg2_detected_cells: Hepg2 detected cells.
k562_detected_cells: K562 detected cells.
hepg2_mean_mpra_umi_per_cell: Hepg2 mean mpra umi per cell.
k562_mean_mpra_umi_per_cell: K562 mean mpra umi per cell.
hepg2_pseudobulk_umi_per_10k_transcriptome: Enhancer UMI total normalized to 10,000 transcriptome UMIs in HepG2-like cells.
k562_pseudobulk_umi_per_10k_transcriptome: Enhancer UMI total normalized to 10,000 transcriptome UMIs in K562-like cells.
sc_log2_activity_HepG2_vs_K562: Log2 ratio of pseudobulk scMPRA activity in HepG2-like versus K562-like cells.
sc_matrix_qc_pass: Pass/fail flag for valid sequence and nonnegative sparse-matrix counts.

Quality control

Kept enhancer features with valid ACGT sequences and nonnegative sparse-matrix counts. The matrix contained 1,345 enhancer rows and 10,640 cells; all 1,345 features passed. HepG2-like/K562-like labels were inferred reproducibly with K-means (random_state=0, n_init=20) on canonical marker modules (5,061/5,579 cells).

Curation notes

Cell-state labels are package-level inferred labels, not an author-provided cell barcode annotation. The GEO enhancer matrix contains 1,345 R1 features, one fewer than the 1,346-row bulk processed table; the unmatched bulk element is not represented in the scMPRA table.

Row
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50

Row
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50

Row
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50